Hi, I am doing research around maternal anxiety. I am trying a new psychological intervention and comparing with a control group. There are previous feasibility studies done by other researchers. But I am trying to figure out if the intervention will have a positive impact on anxiety.
I see that there are small clinical trials with small samples referring to safety and I was wondering if this is the case for this intervention. Can my sample be of 20 (IG=10 and WLC=10)?
Thank you
The problem with small samples is that they how "low power" to detect significant differences, which is due to large standard errors. Another way to think about this is that you need to have a very large effect size for it to make a significant difference in a small number of observations. There is thus a high likelihood of a "Type-Two" error, where an effect is present, but you reject it because you did not have enough observations to detect it at given cut-off level (e.g., p < .05)..
Article How to Calculate Sample Size in Randomized Controlled Trial?
Having Estimates Sample Size,
If the Sample size is small, these are some methods that can be used:
https://www.openaccessjournals.com/articles/small-sample-sizes-in-clinical-trials-a-statisticians-perspective.pdf
السلام عليكم
سيكون التاثير في بعض الاحيان ايجابي وفي الاخرى سلبي حسب نوع القلق
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size